Performance Evaluation of Cloud Services with Profit Optimization

Abstract

Cloud computing is a promising computing paradigm which allows distribution of services from a pool of resources. The services are required by the clients through on-demand via pay and use method. The greatest utilization of resources and maximum profits with scheduling is the main goal of the cloud service providers. The major issue in cloud computing is scheduling of services with improved global throughput and job scheduling. Since, cloud computing is a service based one, the performance evaluation is an important criteria to be dealt with. In this paper, we propose a Priority based Queuing model to evaluate the services leased by the cloud service providers. We consider general service time and response time for arriving requests and the waiting requests are stored in the queue. The services are considered to be SaaS (Software as a Service), PaaS (Platform as a Service) and IaaS (Infrastructure as a Service). We construct this using Queuing model with markovian arrival rate, general service rate, ‘m’ number of servers, priority queue discipline and a buffer of size ‘r’. The advantage of the proposed analytical model is within the time span, the cloud service provider schedules his services to result in maximum profit.